If we remove meta tag "Content-Type" will that affect Google Rank?

Our page says "charset=iso-8859-1" which is Latin?

[JeffOstroff]

Found several of our pages have this meta tag:

<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">

One of our SEO tools indicated this can be removed from our page. But I have always wondered what this line really does and what the effect of delting it will have?

The validator at W3C seems to indicate it's a western Europe character set, but I cannot figure out why.

Even CNN has it on their website too.

Anyone have any insight into this?

[JeffOstroff]

Not sure why evey site seems to have this tag yet SEO software says to remove it, that it's not needed.

[g1smd]

You do not have a Valid HTML document if the document does not declare what character set the page is using. It is a required element.

It should go before the title and meta description tags.

[JeffOstroff]

Our site so far is working wihtout it, so how is this invalid if explorer still shows it properly?

[g1smd]

IE takes a best guess with just about any sort of tag soup that you might throw at it.

Doesn't mean that you should purposely set out to send it a pile of duff code though.

[JeffOstroff]

Got it, so we should specify something. I notice Google uses this code:

<meta HTTP-EQUIV="content-type" CONTENT="text/html; charset=UTF-8">

Why do almost all other pages use the Western Europe one?
<meta http-equiv="content-type" content="text/html; charset=iso-8859-1">

If we ar ein the U.S., I would think UTF-8 should be the choice, but I don't know what the differences are.

FrontPage creates this one when you create a new html page:

<meta http-equiv="Content-Type" content="text/html; charset=windows-1252">

How do we know which one to use?
So now this makes it even more confusing!

[londrum]

UTF-8 is unicode. that is the mother of all character sets, and contains just about every symbol, sign, letter, number and punctuation mark that you can use on the web.

everything else is a subset of it.
iso-8859-1 is the subset that just contains symbols and signs from western languages (english, french, german etc etc)
that is why you see it on just about every site on the web,
- but you could just as easily use the unicode set.

the character set is usually delivered in the header by your web server, which is why SEO's say the meta tag can safely be removed, as its just doubling up what the browser already knows.
but most people leave it on anyway - just in case the page has been saved and is being viewed offline. if it's being viewed offline and you miss out the meta tag then the browser will not know which charcater set is being used (not that it really matters, i don't think)

[g1smd]

It matters a lot if the browser default is different to what the page is actually saved as.

Non-latin alphabet pages (Greek, Cyrillic, Thai, Chinese, Japanese, Arabic) are totally non-viewable in many western browsers (and vice versa) when the character set is missing from the offline page.

[abates]

Is Google even taking any notice of these? I've had Content Type meta tags and a server-set header, both identifying my pages as UTF-8, yet in sitemaps my pages are shown as a mixture of different content types. :P

[JeffOstroff]

I see that too, I see in sitemaps where they show some of my pages as latin, but it does not specify which ones. That is most odd.